Introduction

Comment

It is important to try to establish a diagnosis of autoimmune pancreatitis since treatment consists of steroids. The dilemma arises in making this diagnosis, since clinically and radiologically it can mimic pancreatic adenocarcinoma and traditionally these patients have undergone surgery and the diagnosis made postoperatively. Deshpande et al analyzed patients with presumed AIP based on surgical pathology (11) or clinical findings (5) and concluded that the presence of stromal fragments with high cellularity along with lymphoid infiltration on EUS FNAB may help establish the diagnosis of AIP. This study provides us more information on the cytopathology findings of AIP. When reviewing pancreatic FNAB specimens, in the appropriate clinical setting it may be useful to note the presence/absence of the above findings to aid in diagnosing AIP. Further prospective studies need to be performed to validate these results, but this study provides an important caveat for endosonographers performing EUS FNA and cytopathologists interpreting their aspirates.

Introduction

Chronic pancreatitis (CP) is difficult to diagnose without surgical pathology. Non-operative methods have not been optimal in making this diagnosis. In this study Dewitt et al., examined the utility and safety of EUS-guided trucut biopsy in diagnosing chronic pancreatitis and compared it to EUS and ERCP. Sixteen patients who had >3 EUS features of chronic pancreatitis proceeded to have EUS guided trucut biopsy. Eight had a normal pancreas, probable CP (1), nondiagnostic (6), and device malfunction (1). They reported two complications, acute pancreatitis and persistent abdominal pain requiring 24 hour hospital observation. In the eight pts with CP and normal pancreatic histology by EUS-TCB, ERCP was normal in four, revealed mild or moderate pancreatitis in three, and was not done on one. The sensitivity, the specificity, the positive predictive value, the negative predictive value, and the accuracy of EUS imaging for the diagnosis of CP was 100%, 62%, 64%, 100%, and 78%, respectively. 22 patients underwent both EUS and ERCP, the agreement was fair (kappa, 0.40). When comparing diagnostic EUS and ERCP with EUS-TCB results, agreement was poor (kappa, 0) and fair (kappa, 0.25), respectively. They conclude that because of potential complications and limited diagnostic yield (6/16), trucut biopsy is not currently recommended for use in the routine evaluation of patients with suspected CP.

Comment

EUS guided trucut biopsy may be used to potentially obtain better specimens non-operatively to establish a diagnosis of chronic pancreatitis. Chronic pancreatitis may be difficult to diagnose based on the possibility of a patchy distribution, and the limited data on the true correlation between EUS parencymal/ductal criteria and histological diagnosis. The use of this new device to help diagnose chronic pancreatitis was an innovative concept investigated by the authors in this pilot study. The number of nondiagnostic samples in 6/16 specimens, along with 2 complications, and the poor correlation with ERCP and EUS indicate that trucut biopsy for chronic pancreatitis is not quite ready for “prime-time” yet. Future studies are needed to investigate the safety and the indications of trucut biopsy of the pancreas in chronic pancreatitis, along with modifications of the device to yield better results.

Introduction

Differentiating malignant from non-malignant pancreatic cysts may be difficult. Khalid et al, evaluated the role of molecular analysis of cyst fluid obtained on EUS to help determine malignant potential. Thirty-six patients with histological confirmation were included in the study. Eleven lesions were malignant, 15 were premalignant and 10 were benign. Mean cyst CEA levels for the benign, premalignant, and malignant groups were 13 ng/mL, 5237 ng/mL, and 108,360 ng/mL, respectively. This difference was significant between the benign and premalignant groups (P = .001) and the premalignant and malignant groups (P < .05). Malignant cysts were also differentiated from premalignant cysts by DNA quality (P = .009), and on the sequence of mutations acquired (P < .001). Early k-ras mutation followed by allelic loss was most predictive of malignancy (sensitivity, 91%; specificity, 93%). Six of 15 cysts in the premalignant group and 10/11 malignant cysts carried mutations. No mutations were detected in the benign group. The mean number of mutations in the premalignant group and malignant group were 0.9 and 2.8 respectively. The number of mutations differed significantly between the malignant and premalignant (P = .003), malignant and benign (P < .001), and premalignant and benign categories (P = .036).

Comment

Pancreatic cysts may be difficult to manage preoperatively due to the inability to accurately determine presence of malignancy. EUS, FNA with fluid analysis for cytology and CEA have shown various accuracies in making this differentiation. This study demonstrated the feasibility of DNA extraction from cyst fluid. Cyst fluid analysis for DNA quality, sequence of mutations, number of mutations and CEA appeared to provide us some useful information to help differentiate malignant from pre-malignant and benign lesions. Only a small amount of fluid (0.4ml) was required, but expertise in this complex molecular analysis is needed and probably only available in certain research labs. This study is promising in that advances are being made to help us in differentiating pancreatic cystic neoplasms, as cytology results are variable among different centers. However, further studies need to be performed to assess technique of extraction, reproducibility of results and feasibility to validate the utility of this analysis on a routine basis.